Comparing a new algorithm with the classic methods for estimating the number of factors

被引:74
作者
Henry, RC
Park, ES
Spiegelman, CH
机构
[1] Univ So Calif, Dept Civil & Environm Engn, Environm Engn Program, Los Angeles, CA 90089 USA
[2] Univ Washington, Natl Res Ctr Stat & Environm, Seattle, WA 98195 USA
[3] Texas A&M Univ, Dept Stat, College Stn, TX 77843 USA
基金
美国国家科学基金会;
关键词
number of factors; correlation matrix; eigenvalues; eigenvectors; Bartlett's test; scree plots; Malinowski's indicator function; cross-validation; NUMFACT;
D O I
10.1016/S0169-7439(99)00015-5
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
This paper presents and compares a new algorithm for finding the number of factors in a data analytic model. After we describe the new method, called NUMFACT, we compare it with standard methods for finding the number of factors to use in a model. The standard methods that we compare NUMFACT with are Malinowski's indicator function, Weld's cross-validation approach, Bartlett's test, scree plots, the rule-of-one, and using the number of factors (eigenvectors) needed to explain 90% of the trace of a correlation matrix. Using a diverse set of real applications, NUMFACT is shown to be the clear method of choice. (C) 1999 Elsevier Science B.V. All rights reserved.
引用
收藏
页码:91 / 97
页数:7
相关论文
共 18 条
[1]   ASYMPTOTIC THEORY FOR PRINCIPAL COMPONENT ANALYSIS [J].
ANDERSON, TW .
ANNALS OF MATHEMATICAL STATISTICS, 1963, 34 (01) :122-&
[2]  
[Anonymous], 1969, Bull. Intern. Statist. Inst
[3]  
BARTLETT MS, 1951, BIOMETRIKA, V38, P337, DOI 10.1093/biomet/38.3-4.337
[4]   COMPREHENSIVE TRIAL OF SCREE AND KG CRITERIA FOR DETERMINING NUMBER OF FACTORS [J].
CATTELL, RB ;
VOGELMANN, S .
MULTIVARIATE BEHAVIORAL RESEARCH, 1977, 12 (03) :289-325
[6]  
Friedman JH., 1984, BIOMETRICS, V40, P874, DOI [DOI 10.2307/2530946, 10.2307/2530946]
[7]   Reported emissions of organic gases are not consistent with observations [J].
Henry, RC ;
Spiegelman, CH ;
Collins, JF ;
Park, E .
PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA, 1997, 94 (13) :6596-6599
[8]  
HENRY RC, 1997, 279 TEX A M U
[9]   ANALYSIS OF DAILY PRECIPITATION DATA BY POSITIVE MATRIX FACTORIZATION [J].
JUNTTO, S ;
PAATERO, P .
ENVIRONMETRICS, 1994, 5 (02) :127-144
[10]   ON CLIFF FORMULA, THE KAISER-GUTTMAN RULE, AND THE NUMBER OF FACTORS [J].
KAISER, HF .
PERCEPTUAL AND MOTOR SKILLS, 1992, 74 (02) :595-598